Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS6668309 B2
Publication typeGrant
Application numberUS 10/352,946
Publication dateDec 23, 2003
Filing dateJan 29, 2003
Priority dateDec 29, 1997
Fee statusPaid
Also published asUS6460119, US6578116, US20020199068, US20030115424
Publication number10352946, 352946, US 6668309 B2, US 6668309B2, US-B2-6668309, US6668309 B2, US6668309B2
InventorsDerek T. Bachand, Paul Breuder, Matthew A. Fisch
Original AssigneeIntel Corporation
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Snoop blocking for cache coherency
US 6668309 B2
Abstract
In a method of processing a bus transaction, an address is retrieved from the bus transaction and referred to a queue of pending transaction. A match indicator signal is returned from the queue. If the match indicator signal indicates a match, a snoop probe for the bus transaction is blocked.
Images(4)
Previous page
Next page
Claims(11)
We claim:
1. A transaction management method for a computer system, comprising:
by a first integrated circuit, posting a first transaction on an external bus that requests a copy of addressed data,
by a second integrated circuit:
determining whether the second integrated circuit previously posted on the external bus a second transaction directed to the addressed data,
if so, blocking an internal snoop probe responsive to the first transaction until the second transaction is globally observed, and
after the second transaction is globally observed, issuing the snoop probe and posting on the external bus a snoop response responsive to the first transaction,
by the first integrated circuit, after the snoop response is available on the external bus, changing a cache coherency state associated with the addressed data in a storage location of the first integrated circuit.
2. The transaction management method of claim 1, further comprising following generation of the snoop response, transferring the addressed data from the second agent to the first agent.
3. The transaction management method of claim 1, further comprising following generation of the snoop response, transferring the addressed data from system memory to the first agent.
4. A transaction management method for a computer system, comprising:
by a first integrated circuit, posting a first transaction on an external bus that requests a copy of addressed data,
by a second integrated circuit:
identifying all transactions that are pending on the external bus at the time the first transaction was posted and were posted by the second integrated circuit,
determining whether any such transactions are directed to the same address as the first transaction,
if so, blocking an internal snoop probe responsive to the first transaction until the second transaction is globally observed, and
after the second transaction is globally observed, issuing the snoop probe and posting on the external bus a snoop response responsive to the first transaction,
by the first integrated circuit, after the snoop response is available on the external bus, changing a cache coherency state associated with the addressed data in a storage location of the first integrated circuit.
5. The transaction management method of claim 4, further comprising following generation of the snoop response, transferring the addressed data from the second agent to the first agent.
6. The transaction management method of claim 4, further comprising following generation of the snoop response, transferring the addressed data from system memory to the first agent.
7. An integrated circuit, comprising:
an interface to an external bus,
a transaction queue to manage transactions on the bus, the transaction queue coupled to the interface and comprising a plurality of transaction queue entries to store data of external transactions to be originated by the integrated circuit, and
a snoop queue to manage snoop responses of the integrated circuit in response to external bus transactions, the snoop queue coupled to the interface an comprising a plurality of snoop queue entries, each having a blocking bit field and a pointer to a transaction queue entry, the snoop queue to set the blocking bit field if an address of an external transaction initiated by other integrated circuit on the external bus matches an address of a transaction stored by the transaction queue and pending on the bus.
8. The integrated circuit of claim 7, wherein the external bus is a pipelined bus having a predetermined pipeline depth and the number of snoop queue entries is less than the pipeline depth of the bus.
9. The integrated circuit of claim 7, wherein the transaction queue entries comprise a field to store data associated with a transaction's request type.
10. The integrated circuit of claim 7, wherein the transaction queue entries each comprise a field to store data representing the transaction's stage on the external bus.
11. The integrated circuit of claim 7, wherein the blocking bit field stores data that indicates an internal snoop probe associated with the transaction stored in the respective snoop queue entry is to be blocked from processing.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation application that claims the benefit of U.S. patent application Ser. No. 10/215,599 (filed Aug. 9, 2002) (allowed Nov. 4, 2002) now U.S. Pat. No. 6,578,116; which is a continuation of U.S. patent application Ser. No. 09/210,641 (file Dec. 14, 1998) (allowed Jun. 20, 2002) now U.S. Pat. No. 6,460,119, which is a continuation-in-part of Ser. No. 08/999,242, now U.S. Pat. No. 6,209,068 (filed Dec. 29, 1997), which applications are incorporated herein in their entirety.

BACKGROUND

The present invention relates to a cache coherency technique in an agent using a pipelined bus.

As is known, many modern computing system employ a multi-agent architecture. A typical system is shown in FIG. 1. There, a plurality of agents 10-50 communicate over an external bus 60 according to a predetermined bus protocol. “Agents” may include general purpose processors, chipsets for memory and/or input output devices or other integrated circuits that process data requests. The bus 60 may be a “pipelined” bus in which several transactions may be in progress at once. Each transaction progresses through a plurality of stages but no two transactions are in the same stage at the same time. The transactions complete in order. With some exceptions, transactions generally do not “pass” one another as they progress on the external bus 60.

In a multiple-agent system, two or more agents may have need for data at the same memory location at the same time. The agents 10-50 operate according to cache coherency rules to ensure that each agent 10 uses the most current copy of the data available to the system. According to many cache coherency systems, each time an agent 10 stores a copy of data, it assigns to the copy a state indicating the agent's rights to read and/or modify the data.

For example, the Pentium® Pro processor, commercially available from Intel Corporation, operates according to the “MESI” cache coherency scheme. Each copy of data stored in an agent 10 is assigned one of four states including:

Invalid—Although an agent 10 may have cached a copy of the data, the copy is unavailable to the agent. The agent 10 may neither read nor modify an invalid copy of data.

Shared—The agent 10 stores a copy of data that is valid and possesses the same value as is stored in external memory. An agent 10 may only read data in shared state. Copies of the data may be stored with other agents also in shared state. An agent 10 may not modify data in shared state without first performing an external bus transaction to gain exclusive ownership of the data.

Exclusive—The agent 10 stores a copy of data that is valid and may possess the same value as is stored in external memory. When an agent 10 caches data in exclusive state, it may read and modify the data without an external cache coherency check.

Modified—The agent 10 stores a copy of data that is valid and “dirty.” A copy cached by the agent 10 is more current than the copy stored in external memory. When an agent 10 stores data in modified state, no other agents possess a valid copy of the data.

Agents 10-50 exchange cache coherency messages, called “snoop responses,” during external bus transactions. The snoop responses identify whether other agents possess copies of requested data and, if so, the states in which the other copies are held. For example, when an agent 10 requests data held in modified state by another agent 20, the other agent 20 may provide the data to the requesting agent in an implicit writeback. Ordinarily, data is provided to requesting agents 10 by the external memory 50. The modified data is the most current copy of data available to the system and should be transferred to the requesting agent 10 in response to a data request.

When external bus transactions cause an agent to change the state assigned to a copy of data, state changes occur after snoop responses are globally observed.

As an example, consider a “read for ownership” request issued by an agent 10. Initially, an agent 10 may store the requested data in an invalid state. The agent 10 has a need for the data and issues a bus transaction requesting it. The agent 10 receives snoop responses from other agents 20-40. When the snoop responses are received, the transaction is globally observed. The agent 10 marks the requested data as held in exclusive state. The agent 10 may mark the data even though it has not yet received the requested data. For example, in known processors, data is transferred in a data phase of a transaction following a snoop phase. Before the data is received, an entry of an internal cache (not shown) is reserved for the data. A state field in the external transaction queue is marked as exclusive when the transaction is globally observed and before the requested data is 10 received, but the state field in the reserved cache entry is not marked exclusive until the data is filled into the cache.

Certain boundary conditions arise when state transitions are triggered by the receipt of snoop responses. An example is shown in the following table using the Pentium® Pro bus protocol:

Bus Clocks
1 2 3 4 5 6 7 8 9 10 11
Transaction No. 1 Req Req Err Snoop Stall Snp Resp Data X
State in Agent 10 I I I I I I I E E E E
Transaction No. 2 X X Req Req Err Snoop Stall Snp Resp Data
State in Agent 20 I I I I I I I I E E E

In the boundary condition, without some sort of preventative measure, two different agents 10 and 20 in the system could mark a copy of the same data in exclusive state. To do so would violate cache coherency. Assume that two agents 10 and 20 post read requests to a single piece of data. The first agent 10 posts the request as explained above. When the first transaction concludes its request phase, the second agent 20 posts a second transaction for the same data.

Assume further that the snoop phase of the first transaction is stalled by a snoop stall. A snoop stall signal occurs when an agent (say, agent 30) requires additional time to generate snoop results. Although the first agent 10 may reserve a cache entry for the requested data, the agent 10 does not mark the requested data as exclusive until snoop results for its transaction are received. When snoop results eventually are received for the first transaction (in clock 8), the first agent 10 will mark the data as held in exclusive state. However, the first agent 10 observes the second transaction in clock 3. If it performs internal snoop inquiries for the second transaction before the first transaction is globally observed, its snoop response would indicate that it does not possess a valid copy of the data. The second agent 20 also could mark the data as exclusive. Having two agents 10, 20 each store data in exclusive state violates the MESI cache coherency rules because each agent 10, 20 could modify its copy of the data without notifying the other via a bus transaction.

The coherency violation can arise if an agent 10 begins internal snoop inquiries before its previous transaction to the data is globally observed. Thus, the error can be avoided if the snoop inquiries related to the second transaction are blocked until a prior conflicting transaction related to the same data is globally observed.

The Pentium® Pro processor includes a snoop queue to manage cache coherency and generate snoop responses. The snoop queue buffers all transactions posted on the external bus. For new transactions, the snoop queue compares the address of the new transaction to addresses of transactions that it previously stored to determine whether the addresses match. If so, and if the previous transaction were not globally observed, the snoop queue blocks a snoop probe for the new transaction. The block remains until snoop results for the prior pending transaction are received.

The Pentium® Pro processor's snoop queue is large. The snoop queue possesses a queue entry for as many transactions as can be pending simultaneously on the external bus. It consumes a large area when the Pentium® Pro processor is manufactured as an integrated circuit. In future processors, it will be desirable to increase the pipeline depth of the external bus to increase the number of transactions that may proceed simultaneously thereon. However, increasing the depth of the external bus becomes expensive if it also requires increasing the depth of the snoop queue.

The Pentium® Pro processor's snoop queue fills quickly during operation. The snoop queue buffers not only requests from other agents but also requests posted by the agent to which the snoop queue belongs. Because the Pentium® Pro includes an external transaction queue that monitors transactions issued by the processor, the snoop queue's design is considered sub-optimal.

Accordingly, the inventors perceived a need in the art for a snoop queue in an agent that possesses a depth that is independent of the pipeline depth of the agent's external bus. There is a need in the art for such a snoop queue, however, that maintains cache coherency and insures that, when two bus transactions related to the same address are pending on the external bus at the same time, snoop inquiries related to the second transaction will not be generated until the first transaction has been globally observed.

SUMMARY

Embodiments of the present invention provide a method of processing a bus transaction in which an address is retrieved from the bus transaction and referred to a queue of pending transactions. A match indicator signal is returned from the queue. If the match indicator signal indicates a match, a snoop probe for the bus transaction is blocked.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of a conventional multi-agent system.

FIG. 2 is a block diagram of a bus sequencing unit of an agent constructed in accordance with an embodiment of the present invention.

FIG. 3 is a flow diagram illustrating operation of a snoop queue in accordance with an embodiment of the present invention.

FIG. 4 is a block diagram illustrating relevant portions of an external transaction queue and a snoop queue constructed in accordance with an embodiment of the present invention.

DETAILED DESCRIPTION

The present invention alleviates the disadvantages of the prior art by providing an agent having a snoop queue whose depth is independent of the pipeline depth of its external bus. Embodiments of the present invention provide a snoop queue with a snoop blocking function that is coordinated with an external transaction queue. When the snoop queue observes an external bus transaction, before it issues a snoop probe for cache coherency checks, it refers the address of the new transaction to the external transaction queue. The external transaction queue compares the address of the new transaction with addresses of earlier-posted transactions that have not yet been globally observed. If a match occurs, the external transaction queue identifies the match to the snoop queue which in turn, blocks a snoop probe for the new transaction. After the pending transaction has been globally observed, the block is released.

In an embodiment, the principles of the present invention may be applied in a bus sequencing unit 200 (“BSU”) of an agent, shown in FIG. 2. The BSU 200 includes an arbiter 210, an internal cache 220, an internal transaction queue 230, an external transaction queue 240 and the snoop queue 250. An external bus controller 300 interfaces the BSU 200 to the external bus 60. The BSU 200 fulfills data requests issued by, for example, an agent core 100.

The arbiter 210 receives data requests from not only the core 100 but also from a variety of other sources such as the snoop queue 250. Of the possibly several data requests received simultaneously by the arbiter 210, the arbiter 210 selects and outputs one of them to the remainder of the BSU 200.

The internal cache 220 stores data in several cache entries. It possesses logic responsive to a data request to determine whether the cache 220 stores a valid copy of requested data and, if so, it furnishes the requested data in response thereto.

The internal transaction queue 230 receives and stores data requests issued by the arbiter 210. It coordinates with the internal cache 220 to determine if the requested data “hits” (was furnished by) the internal cache 220. If not, if a data request “misses” the internal cache 220, the internal transaction queue 230 forwards the data request to the external transaction queue 240.

The external transaction queue 240 interprets data requests and generates external bus transactions to fulfill them. The external transaction queue 240 is populated by several queue entries. The external transaction queue 240 manages the agent's transactions as they progress on the external bus 60. For example, when data is available in response to a transaction, the external transaction queue 240 retrieves the data and forwards it to, for example, the core 100.

The snoop queue 250 performs cache coherency checks within the agent. Typically, in response to a new bus transaction issued by another agent, the snoop queue 250 generates snoop probes to various caches within the agent (such as internal cache 220) and to the internal and external transaction queues 230, 240. It receives responses to the snoop probes and generates snoop responses therefrom. If necessary, the snoop queue 250 manages implicit writebacks of modified data from the agent.

The external bus controller 300 drives signals on the external bus as commanded by the external transaction queue 240 and snoop queue 250.

FIG. 3 illustrates a method 1000 of the snoop queue 250 operating in accordance with an embodiment of the present invention. It may begin when another agent requests data in a bus transaction. When a new transaction is posted, the snoop queue 250 decodes the transaction (Step 1010). It determines whether the transaction requires a cache coherency check. If so, the transaction requires a snoop probe (Step 1020). The snoop queue 250 then provides the address of the requested data to the external transaction queue 240 (Step 1030). Based upon a response from the external transaction queue, the snoop queue determines whether the address of the new transaction matches the address of a posted transaction (Step 1040). If so, the snoop queue blocks a snoop probe related to the new transaction (Step 1050).

Eventually, the prior conflicting transaction will be globally observed. When that occurs, the snoop queue releases the block (Step 1060). It emits a snoop probe within the agent and generates a snoop response according to conventional techniques (Step 1070).

If, at Step 1040, no match occurred, the snoop queue 250 advances to Step 1070 and emits the snoop probe.

FIG. 4 is a partial block diagram of the external transaction queue 240 and the snoop queue 250. The external transaction queue 240 is populated by a number of queue entries (“ETQ entries”) 242. For each pending bus transaction posted by the external transaction queue 240, one of the ETQ entries 242 stores information regarding the transaction. Such information may include the request type, the address of the transaction and/or the current phase of the transaction. The address field of each ETQ entry 242 includes match detection logic 244. The external transaction queue also includes observation logic 246 in communication with the match detection logic 244 and with the snoop queue 250.

During operation, the external transaction queue 240 receives an address of a new transaction from the snoop queue 250. The observation detection logic 246 forwards the received address to each match detection logic 244. It also observes outputs of the match detection logic 244 to determine whether the address stored in any ETQ entry 242 matches the received address. In the event of a match, the observation detection logic 246 reads the phase from the matching ETQ entry 242 and determines whether the matching transaction has already been issued onto the bus, but not yet been globally observed. If so, the observation detection logic 246 signals to the snoop queue that a conflict match exists.

The snoop queue 250 is also populated by a plurality of entries (“snoop queue entries”) 252. The number of snoop queue entries 252 is independent of the pipeline depth of the external bus 60. It is also independent of the number of ETQ entries 242. The snoop queue 250 possesses control logic 254 to implement the method of FIG. 3. It forwards the address of new transactions to the external transaction queue 240. The control logic 254 also receives the match signal from the external transaction queue 240. Each snoop queue entry 252 includes a blocking bit (not shown) which, if enabled, prevents the snoop queue 240 from issuing a snoop probe. Responsive to a match signal from the external transaction queue, the control logic 254 enables the blocking bit. The blocking bit remains enabled until the pending conflicting transaction is globally observed. Thereafter, the bit is cleared and a snoop probe may be issued.

In an embodiment, each of the ETQ entries 242 is assigned a unique identifier (“ETQ ID”). When a conflict match exists, the observation detection logic 246 may provide the ETQ ID of the conflicting transaction to the snoop queue 250.

In an embodiment where the external transaction queue 240 furnishes the ETQ ID of a pending conflicting transaction, the snoop queue 240 may store the ETQ ID in a snoop queue entry 252 of the new transaction when it enables the blocking bit. In this embodiment, when the EBC 300 receives snoop responses, it forwards them to both the external transaction queue 240 and the snoop queue 250. The EBC 300 relates the snoop response to a transaction using its ETQ ID. Upon receipt of the snoop responses and the ETQ ID, the snoop queue 250 releases the blocking bit of all snoops which were being blocked by the associated ETQ transaction.

Optionally, the snoop queue 250 may be configured to ignore certain types of transactions. For example, a conflicting write back transaction does not raise coherency issues for a subsequent transaction because global observation of the write transaction does not necessarily mean that the agent is giving up ownership of the cache line. Also, an “uncacheable read,” one that causes an agent to read but not cache requested data, does not cause state changes to occur within the agent when the read transaction is globally observed. In this embodiment, the observation detection logic 246 also reads the request type out of the ETQ entry 242 of the matching pending transaction. Further, a “self snoop,” another transaction identified by its request type, need not block a transaction. The observation logic 246, based on the request type, may not indicate “block” even though an address match occurred with an outstanding transaction.

Thus the present invention provides a snoop queue having a reduced queue size. The snoop queue of the present invention severs the relationship between the depth of the snoop queue and the pipeline depth of the external bus. The snoop queue of the present invention includes a snoop probe blocking feature to eliminate the boundary conditions that may exist when two agent issue transactions requesting the same data.

Several embodiments of the present invention are specifically illustrated and described herein. However, it will be appreciated that modifications and variations of the present invention are covered by the above teachings and within the purview of the appended claims without departing from the spirit and intended scope of the invention.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US5426765 *Apr 13, 1994Jun 20, 1995Compaq Computer CorporationMultiprocessor cache abitration
US5584017 *Apr 18, 1995Dec 10, 1996Intel CorporationCache control which inhibits snoop cycles if processor accessing memory is the only processor allowed to cache the memory location
US5623628 *Mar 2, 1994Apr 22, 1997Intel CorporationComputer system and method for maintaining memory consistency in a pipelined, non-blocking caching bus request queue
US5652859 *Aug 17, 1995Jul 29, 1997Institute For The Development Of Emerging Architectures, L.L.C.Method and apparatus for handling snoops in multiprocessor caches having internal buffer queues
US5694575 *Nov 7, 1995Dec 2, 1997International Business Machines Corp.Direct I/O control system with processor, main memory, and cache employing data consistency retaining scheme
US5737758Apr 4, 1997Apr 7, 1998Intel CorporationIn a computer system
US5737759 *Apr 4, 1997Apr 7, 1998Intel CorporationMethod and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests
US5774700 *Apr 20, 1995Jun 30, 1998Intel CorporationMethod and apparatus for determining the timing of snoop windows in a pipelined bus
US5778438 *Dec 6, 1995Jul 7, 1998Intel CorporationMethod and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests
US5796977 *Jul 29, 1996Aug 18, 1998Intel CorporationHighly pipelined bus architecture
US5802575 *Oct 7, 1997Sep 1, 1998Sun Microsystems, Inc.Hit bit for indicating whether load buffer entries will hit a cache when they reach buffer head
US5875467Apr 4, 1997Feb 23, 1999Intel CorporationMethod and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests
US5890200Apr 4, 1997Mar 30, 1999Intel CorporationMethod and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests
US5893151Apr 4, 1997Apr 6, 1999Intel CorporationMethod and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests
US5923898 *May 14, 1997Jul 13, 1999International Business Machines CorporationSystem for executing I/O request when an I/O request queue entry matches a snoop table entry or executing snoop when not matched
US5991855 *Jul 2, 1997Nov 23, 1999Micron Electronics, Inc.Low latency memory read with concurrent pipe lined snoops
US6035424 *Dec 9, 1996Mar 7, 2000International Business Machines CorporationMethod and apparatus for tracking processing of a command
US6049849 *Apr 14, 1997Apr 11, 2000International Business Machines CorporationImprecise method and system for selecting an alternative cache entry for replacement in response to a conflict between cache operation requests
US6065101 *Jun 12, 1997May 16, 2000International Business Machines CorporationPipelined snooping of multiple L1 cache lines
US6070231 *Dec 2, 1997May 30, 2000Intel CorporationMethod and apparatus for processing memory requests that require coherency transactions
US6078981 *Dec 29, 1997Jun 20, 2000Intel CorporationTransaction stall technique to prevent livelock in multiple-processor systems
US6108735 *Sep 29, 1995Aug 22, 2000Intel CorporationMethod and apparatus for responding to unclaimed bus transactions
US6202125May 6, 1997Mar 13, 2001Intel CorporationProcessor-cache protocol using simple commands to implement a range of cache configurations
US6202506 *May 14, 1997Mar 20, 2001Storck Bicycle GmbhBicycle crank
US6205506 *Aug 25, 1998Mar 20, 2001Stmicroelectronics, Inc.Bus interface unit having multipurpose transaction buffer
US6209068 *Dec 29, 1997Mar 27, 2001Intel CorporationRead line buffer and signaling protocol for processor
US6212590 *Mar 13, 1998Apr 3, 2001Compaq Computer CorporationComputer system having integrated bus bridge design with delayed transaction arbitration mechanism employed within laptop computer docked to expansion base
US6269426 *Jun 24, 1997Jul 31, 2001Sun Microsystems, Inc.Method for operating a non-blocking hierarchical cache throttle
Non-Patent Citations
Reference
1Handy, J., The Cache Memory Book, Second Edition, Chapter 4, pp. 123-186, Academic Press, Inc., San Diego, California, 1998.
2Pentium Pro Family Developer's Manual, vol. 1: Specification, Tabel of Contents Chapters 4.
3Pentium Pro Processor System Architecture; Table of Contents Chapters 7; MindShare, Inc., Tom Shanley.
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US6848032 *Sep 27, 2002Jan 25, 2005Apple Computer, Inc.Pipelining cache-coherence operations in a shared-memory multiprocessing system
US6970962 *May 19, 2003Nov 29, 2005International Business Machines CorporationTransfer request pipeline throttling
US7194567Feb 24, 2005Mar 20, 2007International Business Machines CorporationMethod and system for ordering requests at a bus interface
US7206886Feb 24, 2005Apr 17, 2007International Business Machines CorporationData ordering translation between linear and interleaved domains at a bus interface
US7234017Feb 24, 2005Jun 19, 2007International Business Machines CorporationComputer system architecture for a processor connected to a high speed bus transceiver
US7263585 *Sep 19, 2003Aug 28, 2007Ip-First, LlcStore-induced instruction coherency mechanism
US7275124Feb 24, 2005Sep 25, 2007International Business Machines CorporationMethod and system for controlling forwarding or terminating of a request at a bus interface based on buffer availability
US7275125Feb 24, 2005Sep 25, 2007International Business Machines CorporationPipeline bit handling circuit and method for a bus bridge
US7330925Feb 24, 2005Feb 12, 2008International Business Machines CorporationTransaction flow control mechanism for a bus bridge
US7392351Mar 29, 2005Jun 24, 2008International Business Machines CorporationMethod and apparatus for filtering snoop requests using stream registers
US7469312Feb 24, 2005Dec 23, 2008International Business Machines CorporationComputer system bus bridge
US7590805 *Dec 29, 2005Sep 15, 2009Intel CorporationMonitor implementation in a multicore processor with inclusive LLC
US7757032Aug 20, 2008Jul 13, 2010International Business Machines CorporationComputer system bus bridge
US7987322Jan 3, 2008Jul 26, 2011Freescale Semiconductor, Inc.Snoop request management in a data processing system
US8099560Aug 29, 2008Jan 17, 2012Freescale Semiconductor, Inc.Synchronization mechanism for use with a snoop queue
US8131947Aug 29, 2008Mar 6, 2012Freescale Semiconductor, Inc.Cache snoop limiting within a multiple master data processing system
US8131948Aug 29, 2008Mar 6, 2012Freescale Semiconductor, Inc.Snoop request arbitration in a data processing system
US8135917Jun 11, 2008Mar 13, 2012International Business Machines CorporationMethod and apparatus for filtering snoop requests using stream registers
US8200908Feb 6, 2009Jun 12, 2012Freescale Semiconductor, Inc.Method for debugger initiated coherency transactions using a shared coherency manager
US8255638May 1, 2008Aug 28, 2012International Business Machines CorporationSnoop filter for filtering snoop requests
US8327082Aug 29, 2008Dec 4, 2012Freescale Semiconductor, Inc.Snoop request arbitration in a data processing system
US8677073Aug 16, 2012Mar 18, 2014Intel CorporationSnoop filter for filtering snoop requests
US8688910Feb 6, 2009Apr 1, 2014Freescale Semiconductor, Inc.Debug control for snoop operations in a multiprocessor system and method thereof
US20100010965 *Jul 8, 2008Jan 14, 2010International Business Machines CorporationQuery Management Systems
US20140245273 *Dec 29, 2011Aug 28, 2014Shlomo RaikinProtecting the integrity of binary translated code
WO2006104747A2 *Mar 17, 2006Oct 5, 2006Matthias A BlumrichMethod and apparatus for filtering snoop requests using stream registers
Classifications
U.S. Classification711/146, 711/140, 711/152, 711/E12.033, 711/141
International ClassificationG06F12/08
Cooperative ClassificationG06F12/0831
European ClassificationG06F12/08B4P4
Legal Events
DateCodeEventDescription
May 14, 2014ASAssignment
Owner name: SONY CORPORATION OF AMERICA, NEW YORK
Effective date: 20140402
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:INTEL CORPORATION;REEL/FRAME:032893/0199
Jun 15, 2011FPAYFee payment
Year of fee payment: 8
Jun 22, 2007FPAYFee payment
Year of fee payment: 4